Investors pay a lot of attention to market valuation, for good reasons. Understanding how expensive or cheap stocks are relative to earnings (the P/E ratio) has been one of cornerstones of active investing. Peter Lynch used to say “corporate earnings drive stock prices.” Investors often look at expected forward earnings as part of trying to improve their return, and/or to protect their portfolio against a significant market downturn.

There is some confusion about what earnings per share, or EPS, really are for the U.S. stock market. There are three common measures of company earnings, each used to calculate the total aggregate EPS for the S&P 500 index, both actual (trailing), and expected (forward).

1. Net earnings: reports net income for each company based on the Generally Accepted Accounting Principles (GAAP).
2. Operating earnings: exclude “extraordinary items” – income and expenses that are typically considered non-recurring for most businesses according to industry practices. Because mostly expenses (not income) are excluded, these EPS are typically higher than (1)
3. Adjusted earnings: exclude extraordinary items that are considered non-recurring by the company. Because it’s the company’s decision, this number tends to be even higher than (2), the highest of the three numbers.

Net EPS is the true GAAP income that all public companies are legally obligated to report, and is almost always the lowest of the three. The idea behind adjustments was to remove volatile, one-time items in order to arrive at EPS from continuing operations. In practice, adjustments make EPS look better, and the more of them, the higher the number looks. Adjusted EPS (3) is the “headline number” that Wall Street analysts estimate, and the financial media follow when they cover whether the company beat expectations.

All three numbers are freely available online. Standard & Poor’s maintains a spreadsheet with Net (1) and Operating (2) EPS for total index and by sector, for the S&P 500 large-cap, S&P 400 mid-cap, S&P 600 small-cap and S&P 1500 total market. It’s a treasure trove of information about past earnings, but their forward estimates for 2016-2017 may be unreliable (see below). FactSet provides Adjusted EPS (3) for the S&P 500 in a weekly report.

Below are actual trailing 12-month (Q1-Q4 2015) S&P 500 EPS compiled by Standard & Poor’s, and the corresponding P/E Ratios. It’s remarkable how much the “GAAP gap” between Net and Operating EPS widened recently. Net income fell much more in 2015 than Operating, primarily due to losses in Energy and Materials sectors.

The chart below shows a 20-year history of trailing 12-month EPS (rolling quarterly). This is Operating EPS, the middle of the three reporting methods. We can see that earnings dropped by 11% in 2015 – the first annual earnings decline since 2009.

Source: Standard & Poor’s.

When it comes to forward earnings estimates, it gets a bit more complicated. An estimate for a given quarter depends on how far in advance we look at it – it follows a “J”-curve pattern as the quarter approaches. Estimates far in advance of the quarter, created by Wall Street analysts, tend to be overly optimistic (the high tip of the “J”). As the quarter ends, companies provide guidance that reflects where their number will actually be, and also leave themselves room to beat it. Finally, companies “beat the number” by an average of 3-4% during the earnings season (the lower tip of the “J”). But by this time, people focus more on the next quarter rather than the past. This explains why far-in-advance forward EPS estimates, like Shangri-La, are often so attractive but ever-elusive in reality.

Adjusted EPS (the most optimistic) estimate for the calendar year 2016 is around $125, according to FactSet, which would imply forward P/E ratio of 16.4. S&P’s forward Operating EPS is currently $118.1, which implies forward P/E ratio of 17.3. These far-ahead forward estimates are always overstated to some extent – they represent the high tip of the J curve. And in the current environment of declining earnings, due to significant headwinds of slow economic growth and deflation, these projections may be even less reliable.

Trailing P/E ratio is our best way to assess the stock market valuation, in my view, because it doesn’t have the estimate error and upward bias of ever-optimistic analyst consensus. This is especially true in this environment of still-optimistic forecasts in the face of weak fundamentals. Again, the TTM P/E Ratio is 20.5, compared to historical average of 16.4 on this basis, and close to the highest in 12 years:

The only realistic forward view is always (and especially now) one quarter ahead. Q1-2016 earnings season is about to begin. Adjusted Q1 EPS are now expected to decline by 8% YoY, according to Factset – but remember, we’re in the low point of the “J.” If companies “beat the number” in Q1 by the same margin as average, 3-4%, Q1 EPS will likely decline by around 4% when all reported. This gives us Operating EPS of $99.4 for 12-month period from Q2 2015 to Q1 2016.

In the current environment of very high valuation, it’s hard to imagine how stocks can climb much further in the mid term, unless earnings growth resumes. And in the short term, the risk of a sharp correction is high as Q1 earnings get into focus and forward estimates are adjusted.

Model Capital Management LLC is a tactical investment manager. For details about Model Capital’s fundamental, forward-looking approach to tactical investment management, visit here; review the list of our tactical management strategies here. Please contact us for information on how we work with advisors.